[maintainability] QuantumManager::circuit()/batch() re-implement driverAlias()'s default-resolution logic instead of calling it

[corgab]

Summary

QuantumManager::driverAlias(string|UnitEnum|null $driver) already centralizes "resolve $driver to a string, falling back to getDefaultDriver() when null, and unwrapping enums" (src/QuantumManager.php:52-61), and is used by driver() for exactly this purpose. circuit(?string $driver = null) and batch(array $circuits, ?string $driver = null) each independently re-implement the null-fallback half of that same logic inline as $driver ?? $this->getDefaultDriver(), rather than delegating to driverAlias().

Where

• src/QuantumManager.php:52-61 (driverAlias() — the existing centralized logic)
• src/QuantumManager.php:69-75 (circuit() — line 74: $driver ?? $this->getDefaultDriver())
• src/QuantumManager.php:80-89 (batch() — line 87: $driver ?? $this->getDefaultDriver())

Note: because circuit()/batch() type their own $driver parameter as ?string (not string|UnitEnum|null like driver()'s), the enum-unwrapping half of driverAlias() is not reachable through these two methods today — only the null-coalescing half is actually duplicated.

Refactor proposal

Have circuit() and batch() call $this->driverAlias($driver) instead of repeating $driver ?? $this->getDefaultDriver() inline, so there is exactly one place that knows how an absent driver name resolves to the default.

Severity: low